Explanatory alphabetical list of Romano-British place-names
Part 2: B
Part 2a: Ba to Bo
[For information as to which names are included in this list and which not, and an explanation of the abbreviations employed, click on Alphabetical List menu provided above]
BANNA
|
(Celt) |
(Rav) |
|
|
|
(Mod) |
|
Banva |
Banna (151) |
|
|
|
Throp |
|
|
|
|
|
|
(Cumbria) |
Note 1:
This name is conventionally considered to be derived from a hypothetical root banno or banna, taken to mean 'peak, horn'.The name is also found as an element of Gobannio (Abergavenny) and Bannovalum (most probably at Caistor-on-the-Wolds), and what all of these places have in common is not that they are associated with a peak or horn, but that they are built on a high slope, where this may mean either that the fort is actually built on a slope, with one side of the fort higher than the opposite side, or that it is built on a terrace jutting out of a high slope. It is thus clear that the original form of Banna was Banva, the change from v to n being common in Romano-British names. Banva, meaning 'high hill slope', is one of a family of topographical elements in the various hill-letters - thus one sees brav in Bravnogenium (the original form of Ptolemy's Brannogenium at Leintwardine) and bamv in Bamvocalia (the original form of Ravenna's Pampocalia at Ilkley).
It has been fashionable for some forty years to assign Banna to Birdoswald, but since the name will originally have been Banva, meaning 'high hill slope', it is clearly inappropriate for Birdoswald. The Ravenna list has come along from Wallsend (Serduno) to Haltwhistle Burn (Esica), has already mentioned Carvoran (Magnis) and Birdoswald (Gabaglanda), and then proceeds with Uxelludamo (Castlesteads or Old Church Brampton, depending on the date of the list). Banna, between Esica and Uxelludamo in the list can only have been Throp, since Bewcastle to the north is too far away to be included in this list of names. And Banva at Throp was indeed built on a slope.
Note 2:
Example of anva → anna
[The entry for Banna was last modified on 27 February 2021]
BANNATIA
|
(Celt) |
|
(Ptol) |
|
|
|
|
Bannata |
|
Bannatia |
|
|
|
|
or |
|
|
|
|
|
|
Bannatis |
|
|
|
|
|
Note:
Ptolemy places Bannatia in the territory of the Vacomagi. Rivet and Smith see the name as being derived from hypothetical banno or banna, taken to mean 'peak, horn', together with a suffix of apparently unknown meaning.
Ptolemy’s Bannatia and Tamia, and also his Lindum in the territory of the Damnoni, might all be names referring to places in the territory of the Vacomagi. It would appear that this tribe used the hill-letter n2, this hill-letter being seen in the place-name Pinnatis (Inchtuthil) and the corresponding river-letter m being seen in the river-name Veromo, the old name of the river Almond, which joins the Tay at the Roman fort now known as Bertha. This fort was Roman Tuessis. Ptolemy indicates that Tuessis (spellt Tuesis in his Geography) and Pinnata Castra were in the territory of the Vacomagi. But Bannatia and Pinnata are in fact the same Celtic name, the intial B changing to P in the Roman form of the name, the change in the first vowel and in the ending of the name being of no consequence. The old-style Ban element means ‘high hill’ and the inversion-type element nat means ‘hill high’. Scholars normally assume that Ptolemy did not give the names of native settlements, but here we may see an exception. Ptolemy may have given the name of a Celtic settlement called Bannatia and the Romans may have transferred this name to their fortress at Inchtuthil, the name changing at some time to Pinnata Castra (Pinnatis in Ravenna). In the entry for Pinnatis in the Alphabetical List the writer suggests that the Celtic settlement was at Dunkeld. That may well be so, but more specifically it may have been on the site later occupied by the "Pictish" fort now known as the King’s Seat, on the western side of Dunkeld. The high hill referred to in the name may be not the low hill on which the fort stood, but the much higher hill immediately to the north. The territory of the Vacomagi may have extended to the north of Dunkeld and it is here that Ptolemy’s Lindum and Tamia come into play, for Lindum is a compound place-name in the hill-letters l1 and n2 and Tamia is the corresponding river-name (cf. Londinium on the Tamesis and Lindum Colonia on the Witham). In other words Ptolemy’s Tamia is actually a river-name, where m is the river-letter applied to main rivers by people who used the hill-letter n2, in this case by the Vacomagi. If Bannatia was a native settlement then the same might be true of Lindum. It will have been a settlement at the top of a hill close to the river Tay, in the territory of the Vacomagi rather than that of the Damnoni (where Ptolemy places it). But the Celts are not likely to have transferred a river-name, in this case Tamia, to a settlement near the river, whereas the Romans did this on a number of occasions, for example Derbentione, Isurium, Matovion, Rumabo and Veromo. Ptolemy’s Tamia will therefore have been a Roman fort close to the river Tay somewhere upstream from Dunkeld. A location close to the confluence of the Tummel and Tay would seem most likely since this would enable the occupants of the fort to control both valleys. The valley of the Tummel will have been of great interest to the Romans (still planning to conquer the whole island of Britain) since it provided a route to the north, just as today that is the route taken by the railway line from Perth to Inverness and by the A9 trunk road. But a Roman fort near the confluence of the Tummel and Tay (no trace of such a fort has yet been found, so far as the present writer is aware) is likely to have been short-lived. It was probably built in AD85, the year work appears to have started on the legionary fortress at Inchtuthil, and it was presumably abandoned at the same time as that fortress, this being thought to have occurred in AD86 or 87.
Note that the argument set out above implies that the Tay was called Tamia upstream from its confluence with the Almond, but Tavus downstream. This need not be a problem, since there is no good reason to object to one and the same river having different names in the territories of different tribes, more especially if the tribes spoke different languages. The region around the lower reaches of the Tay appears to have been at one time in the territory of a tribe which used the hill-letter s and the river-letter b. One sees the s in the name Lecimocsava (Ravenna’s Levioxava) of the hillfort now known as Castle Law, Abernethy. The part of the river Tay in that region may then have had a name such as Abus. Part of that region later fell into the hands of people who used the hill-letter l1 (but not Abernethy, the l in Lecimocsava being l2) and coined inversion-type place-names. It was these people who put the T in Tabus, this name later becoming Tavus with the common change of b to v.
Lindum is assumed above to be the name of a Celtic settlement, but as given by Ptolemy it may in fact be the name of a Roman fort to which the name of the settlement had been transferred. That fort is most likely to have been close to the river Tay at some point near the eastern end of Loch Tay, though no trace of such a fort has yet been found, so far as the present writer is aware.
[The entry for Bannatia was last modified on 27 February 2021]
BANNAVENTA
|
(Celt) |
|
|
(AI) |
|
(Mod) |
|
possibly Banvaventa |
|
|
Bannaventa |
|
Whilton Lodge |
|
but more probably |
|
|
(Iter II) |
|
(Northamptonshire) |
|
Bandaventena |
|
|
Isannavantia |
|
|
|
(a river) |
|
|
(Iter VI) |
|
|
|
|
|
|
Bannavanto |
|
|
|
|
|
|
(Iter VIII) |
|
|
Note 1:
Rivet and Smith regard Bannaventa as being the correct form. They quote the conventional derivation of the banna element from hypothetical banno or banna, taken to mean 'peak, horn'. As to venta they quote the suggestion of Williams that this was a Celtic word meaning 'field', perhaps with a secondary meaning 'market place'. Bannaventa might then mean 'prominent field' or 'market on the spur'. But Rivet and Smith appear to lean more towards proposals of Rostaing, who wrote of hypothetical pre-Indo-European vin, meaning 'hill', and ben, meaning 'height, peak', so that Bannaventa could be taken to mean 'spur-hill' or 'promontory-hill'.
But Bannaventa may originally have been Banvaventa, where banv is an old-style element meaning 'high hill slope' and venta an inversion-type element meaning 'slope of hill high'. Each element of the name is a mirror image of the other, bearing in mind that 'high' in old-style names is represented by b and in inversion-type names by t. Note that the Romano-British walled town did actually stand on a slope.
It is more probable, however, that the original form of Bannaventa was Bandaventena (though the ending need not have been ena), this being a river-name of the kind having a river-suffix, here vent(ena), attached to a place-name, here Band(ion), comprising one or more hill-letters. The river will be the stream flowing down the eastern side of Borough Hill, immediately east of Daventry in Northamptonshire. Bandion, where the element Band means ‘high hill summit’, will have been the hill-fort (or one of the hillforts) on the summit of Borough Hill, the hill-letter n being n1 since the hillfort is thought to date back to at least the 3rd century BC (see, for example, the Pastscape website of Historic England). The river-suffix ventena comprises the river-letters b (changed to v) corresponding to the hill-letter s and t corresponding to the hill-letter l. The new town at Whilton Lodge was close to the river and the Romans simply transferred the name of the river (with the known nd→nn change) to the town.
Note that it is possible that two thousand years ago that stream was regarded as the main headwater of the river Nene and its name Bandaventena was accepted as the name of the river further downstream. That being so it is probable that the modern river-name Nene is derived from the river-suffix ventena. Perhaps the initial v changed to n and the nen part of the modified suffix developed to form the modern name Nene. Or perhaps the t was lost or omitted and the nena part of the suffix changed at some stage to Nene.
Note, too, that it may be more than coincidence that the modern town-name Daventry includes the davent part of Bandaventena. Perhaps Bannaventa was the official Roman name for the town at Whilton Lodge but the local population used the form Bandaventa. The latter form, or the memory of it, may have survived long enough for it to be transferred to a new settlement founded on the site of modern Daventry, though one would still have to explain the ry at the end of Daventry.
Note 2:
Example of nd → nn
[The entry for Bannaventa was last modified on 27 February 2021]
BANNIO
|
(Celt) |
(Rav) |
|
(AI) |
|
(Mod) |
|
(Go)banvio |
Bannio (53) |
|
Gobannio |
|
Abergavenny |
|
|
|
|
(Iter XII) |
|
(Monmouthshire) |
Note 1:
Rivet and Smith refer to Roman iron-workings in the neighbourhood and see the name as being actually the then name of the river Gefenni, this being derived from hypothetical Gobannia taken to mean 'river of the blacksmiths' or 'river of the ironworks', the name then being transferred from the river to the Roman fort.
However, the Roman fort was built on an area of level ground on an otherwise steep, high slope, so the name appears to be a straightforward topographical name where bann was originally banv meaning 'high hill slope', and the initial Go in the AI form adds the information that the slope was considered steep as well as high. It is not clear whether the original Celtic name included initial Go and this was omitted in the Ravenna form, or whether the Go was added later than the date of the 1st century map used by the compiler of Ravenna.
Note 2:
Possible example of omission or addition of initial letter - G
Example of nv → nn
BANNOVALUM
|
(Celt) |
(Rav) |
|
|
|
(Mod) |
|
Banvobalum |
Bannovalum (105) |
|
|
|
Caistor-on-the-Wolds |
|
|
|
|
|
|
(Lincolnshire) |
Note 1:
The first element of this name is conventionally taken to be derived from hypothetical British banno or banna taken to mean 'peak, horn', and the second from hypothetical val/valio taken to mean 'strong', the name as a whole thus perhaps meaning 'strong spur, prominent height'.
In several names in Roman-Britain, however, the element bann was originally the old-style element banv meaning 'high hill slope'. The element val is an inversion-type element meaning 'slope of hill'. But the elements are in the wrong order within the name, so val must originally have been the old-style element bal meaning 'high hill'. The name is much more appropriate for Caistor-on-the-Wolds than it is for Horncastle, since Caistor actually stands on the western escarpment of the Wolds, the top end of the modern town being some 80 metres above the bottom end. Other examples of banv → bann can be seen in Banna and Gobannio.
Note 2:
Example of nv → nn
Example of b → v
[The entry for Bannovalum was last modified on 27 February 2021]
BEGESSE
|
(Celt) |
(Rav) |
|
|
|
(Mod) |
|
Becsesse |
Begesse (194) |
|
|
|
Seabegs |
|
|
|
|
|
|
(Falkirk) |
Note 1:
Williams suggested derivation from a hypothetical Indo-European root bheg, taken to mean 'bend, curve' or 'ridge'.
However, Begesse appears to be a normal topographical name of the kind having an essa-type ending, but the hill-letter s is missing. The Celtic name will have been Becsesse, where the Becs element means ‘high steep hill’. The hill-letter s corresponds to the river-letter b, and this is seen, with the b changed to v, in the modern river-name Avon. The Becs element is seen again in the preceding Ravenna name, Pexa, which will have been Celtic Becsa. These are old-style names in the hill-letter s and in all probability refer to hillforts, as in the case of Lecimocsava (Ravenna’s Levioxava) at Castle Law, Abernethy, and also Marcotacson (Ravenna’s Marcotaxon) at Castle Craig, south of Pairney, both in Perth and Kinross. It thus seems reasonably clear that Celtic Becsesse and Becsa were respectively the hill-forts on Bowden Hill (southwest of Linlithgow) and Cockleroy (south of Linlithgow), the esse ending of Becsesse indicating that the hill-fort was at the top of a steep slope and overlooked a river, as is indeed the case here, the river being of course the Avon. Note that the two hill-forts are just a little over one kilometre apart. It is probable that Cockleroy, on the highest summit in that area, is the earlier of the two hill-forts. It probably became too cramped at some stage and as there was no room for expansion its occupants decided to build a second hillfort on the nearby and slightly lower summit of Bowden Hill, applying to the second fort the name Becsa of their first fort, but with an essa-type ending to distinguish the two forts from one another.
The Romans transferred the names of both hillforts (and presumably the inhabitants of the forts as well) to new fortlets which they built as part of the preparations for the construction of the Antonine Wall across the Forth-Clyde isthmus. It would appear that the name Becsa, modified to Pexa, was applied to a fortlet at Mumrills, and that Becsesse, modified to Begesse, was applied to a fortlet at Seabegs. For further discussion of Roman Pexa and Begesse in the context of the Antonine frontier see Chapter 22 (The Antonine Wall) of the Home menu.
Note 2:
Example of missing letter - s
[The entry for Begesse was last modified on 27 September 2020]
BEREDA
|
(Celt) |
(Rav) |
|
(AI) |
|
(Mod) |
|
Bereda |
Bereda (128) |
|
Voreda |
|
Plumpton Wall |
|
|
|
|
(Iter II) |
|
(Cumbria) |
Note 1:
Jackson argues for a derivation from hypothetical British Voreda meaning 'horse stream' based on hypothetical voreda, given the meaning 'horse'. Williams also gives this derivation but mentions also an alternative meaning, namely 'path, way'.
The plain fact is, however, that the fort stands at the top of a high slope on the eastern side of the river Petteril, so that Ravenna's form is topographically correct as it stands, bered being an old-style element meaning 'high hill summit'.
In the AI form the initial B of the original name has changed to V and there is an insignificant change in the first vowel of the name.
Note 2:
Example of b → v
BINDOGLADIA
|
(Celt) |
(Rav) |
|
(AI) |
|
(Mod) |
|
Bindogladia |
Bindogladia (38) |
|
Vindogladia |
|
Weatherby Castle |
|
|
|
|
(Iter XII) |
|
(Dorset) |
|
|
|
|
Vindocladia |
|
|
|
|
|
|
(Iter XV) |
|
|
Note 1:
Rivet and Smith see Vindocladia as being the correct form, the first element being derived from hypothetical British vindo taken to mean 'white', 'bright, fair' or 'happy, fortunate'. The second element is in their view derived from hypothetical British clad meaning 'to dig' or clado meaning 'ditch'. The name as a whole is thus taken to mean 'white ditches', referring to defensive ditches cut in chalk.
The earliest form we have of this name is, however, Ravenna's Bindogladia, and this is a straightforward topographical compound in which the old-style element glad meaning 'steep hill summit' is qualified by the earlier old-style element bind meaning 'high hill summit'. The name as a whole thus refers to a location at the top of a steep, high hill, a description which is entirely appropriate for the hill-fort known as Weatherby Castle. Note that the name actually indicates that the hillfort was originally called Bindo and was at some point taken over by the Durotriges, who used the hill-letter l1 (see 'Ptolemy's Celtic tribes in Britain'). The AI forms presumably refer to a posting station at the foot of the steep, high hill, on the road from Badbury Rings to Dorchester. The posting station will then simply have taken its name from the old hill-fort.
The two AI forms show the change of b to v.
Note 2:
Example of b → v
BLATOBULGIO
|
(Celt) |
|
|
(AI) |
|
(Mod) |
|
Bladobulgio |
|
|
Blatobulgio |
|
Birrens |
|
|
|
|
(Iter II) |
|
(Dumfries and Galloway) |
Note 1:
Jackson took the view that the original name was British Blatobolgion, the first element of which was hypothetical blato meaning 'bloom, blossom' or hypothetical blato derived from earlier hypothetical mlato taken to mean 'flour'. The second element he saw as being hypothetical bolgo meaning 'bag, bulge'. The sense of the name would thus be 'flowery hillock' or 'flowery hollow', or better, Jackson thought, 'flour sack', a sort of nickname given to the fort since it had three granaries.
The above seems very far-fetched. It is much more likely that Blatobulgio is simply a slightly modified version of the original Celtic name, for with the very minor change of t to d one obtains Bladobulgio, and this is a straightforward topographical compound with the old-style element blad meaning 'high hill summit' and the transitional element bulg meaning 'high hill steep'. The name as a whole thus simply refers to the location of the fort at the top of the high, steep river embankment at Birrens.
But note that the name would be entirely appropriate for the Iron Age hill-fort on Burnswark Hill and may simply have been transferred by the Romans from the hill-fort to their own fort at Birrens. The distance between the hill-fort and Roman fort, some five kilometres, may speak against this, but the possibility cannot be ruled out. Indeed, the structure of the name - old-style element plus transitional element - indicates that the settlement concerned was founded well before the Romans came to Britain, so that if there is no evidence of an earlier Celtic settlement underneath or adjacent the Birrens fort, then Bladobulgio most probably was the hill-fort on Burnswark Hill. This assumes, of course, that the place-name was indeed the name of an Iron Age settlement and was not merely a topographical name applied by the locals to a feature of the landscape, e.g. the high, steep river embankment at Birrens.
Note 2:
Example showing d → t
[The entry for Blatobulgio was last modified on 20 May 2021]
BLESTIO
|
(Celt) |
|
|
(AI) |
|
(Mod) |
|
Velesedio |
|
|
Blestio |
|
Monmouth |
|
or |
|
|
(Iter XIII) |
|
|
|
Velestio |
|
|
|
|
|
Note 1:
Jackson thought the name could possibly include a personal name Blestus, the place-name perhaps meaning 'Blestus's place'.
But Blestio is a topographical name, though one which has been altered. It will originally have been somewhat of the form Velesedio and was the name of the hill-fort now known as Little Doward Camp, located at SO 539 160, some 5 kilometres to the northeast of Monmouth. This hill-fort stands at the top of a high hill and comprises two enclosures, the NW enclosure extending from the summit, at about 220 metres, down the slope to about the 190 metre contour. The SE enclosure starts at this point and extends further down the slope to about the 165 metre contour. The original name, that of the NW enclosure, will have been sedio, where sed means ‘hill summit’. Much later the hill-fort was taken over by people who used the hill-letter l. They built the SE enclosure and gave it the name-element vel, meaning ‘on the slope of a hill’. The full name of the place, Velesedio, thus comprises the inversion-type generic term vel qualified by the earlier name of the hill-fort, i.e. by sedio. The Celtic name Velesedio was then transferred by the Romans to a fort which they built at Monmouth, but with the common changes v → b and d → t, and Belesetio shortened to Blestio. Note that the vel element will have been added by the Silures – they used the hill-letter l1 (see ‘Ptolemy’s Celtic tribes in Britain’). But note further that if that area was settled by s-people at some time during the second half of the second century BC, or even a little later, then the Celtic name may have been Velestio, where st is an inversion-type element meaning 'hill high'. One sees the st element further north in Argistillum. The Romans appear to have applied this name to their fort at Stretton Grandison, but it will earlier have been the name of a Celtic hillfort/settlement somewhere in the vicinity.
Note 2:
Example of v → b
Possible example of d → t
[The entry for Blestio was last modified on 9 November 2019]
BOGRANDIUM
|
(Celt) |
(Rav) |
|
|
|
(Mod) |
|
Bograndium |
Bograndium (218) |
|
|
|
Ardoch |
|
|
|
|
|
|
(Perth and Kinross) |
Note:
Holder wrote of a hypothetical boc which was taken to mean either 'swelling' or 'soft, tender'. The second element of Bograndium (or Bocrandium, as Rivet and Smith prefer to spell it) is taken by Rivet and Smith to correspond to a hypothetical Gaulish word randa or rando-s given the meaning 'parcel, part'. Bocrandium might then mean 'hilly part' or, following a suggestion by Williams, 'mossy part'.
Bograndium appears, however, to be a straightforward topographical compound of the old-style element bogr meaning 'high steep hill' and old-style and meaning 'hill summit'. The hill-letter n of the and element would then presumably refer to the river embankment on the western side of the fort and/or the embankment on the southern side of the fort. It is more likely, however, that Bograndium was the name of the native fort on Grinnan Hill in the southwestern outskirts of the village of Braco and the name was simply transferred by the Romans to their fort on the northeastern side of the village. Note that the Roman fort is conventionally given the name Alavna, it being assumed that it took its name from the then name of the Allan Water. But the Ardoch fort does not stand on the Allan Water. It is on the river Kraik, which flows down the western side of the fort. We can therefore be quite sure that the fort was not called Alavna.
BOLVELAUNIO
|
(Celt) |
(Rav) |
|
|
|
(Mod) |
|
Bolvelavnio |
Bolvelaunio (31) |
|
|
|
Wiveliscombe |
|
(a river) |
|
|
|
|
(Somerset) |
|
or |
|
|
|
|
|
|
Bolbelagunio |
|
|
|
|
|
|
or |
|
|
|
|
|
|
Borbeladunio |
|
|
|
|
|
Note:
Williams saw the bol element as being derived from hypothetical bol, bul, taken to mean 'hill'. The remainder of the name is thought to be derived from hypothetical vellauno which, following a suggestion of Holder, is taken to mean 'good'.
Bolvelavnio appears, however, to be a river-name of the kind comprising a river-suffix, here elavnio, attached to a place-name comprising one or more hill-letters, here the old-style element bolv, which means 'high hill slope'. The fort at Wiveliscombe was indeed built on a hillside, a little to the SE of the modern town. The hill rises to a height of some 180 metres above the fort itself, so the element bolv appears entirely appropriate. Nonetheless it should be noted that the place called Bolv will have been a Celtic settlement, and it need not have been on exactly the same site as the Roman fort. The river-suffix elavnio comprises the river-letters l and b, changed to v.
The above explanation may be correct but is not entirely satisfactory since the river-letter b corresponds to the hill-letter s and this hill-letter was in use in this region earlier than the hill-letter l (cf. Celtic Ucsela/Ucselda/Ucselva - Ptolemy's Uxella - at Cannington Park near Combwich). Perhaps, then, one should not expect to see the river-letter b in a suffix attached to an old-style name in the hill-letter l. It seems much more likely that Bolvelaunio is a corruption of the Celtic name of the Castle Hill hill-fort, standing at the top of a steep, high hill immediately east of Wiveliscombe. The Celtic name may have been Bolbelagunio, similar to Bladobulgio, apparently the Celtic form of the Blatobulgio of the Antonine Itinerary (at Burnswark Hill/Birrens). The old-style Bol element means 'high hill' and the transitional element belag means 'high hill steep'. Alternatively, the second element may have been old-style belad (similar to the first element of Bladobulgio) meaning 'high hill summit'. The first element will also have been an old-style element but in a different hill-letter, perhaps r, this yielding a form Borbeladunio. Either form of the name would be appropriate for the site of the hill-fort. The name will then have been transferred by the Romans to the fort which they built a little to the south of the hill-fort. If the Celtic name had been Bolbelagunio then at some stage the second b changed to v and the g was dropped, thus yielding Ravenna's Bolvelaunio. If the Celtic name had been Borbeladunio then the r was at some stage replaced by l, the second b changed to v and the d was dropped.
It seems most likely that the Wivel element of the modern name Wiveliscombe is derived from the Bolvel element of Bolvelaunio, though English scholars prefer a derivation from a personal name Wifela or Wyvel.
BOMIO
|
(Celt) |
|
|
(AI) |
|
(Mod) |
|
Bomio |
|
|
Bomio |
|
Cowbridge |
|
|
|
|
(Iter XII) |
|
(Glamorgan) |
Note:
Rivet and Smith amend this name to Bovium and see a derivation from hypothetical British bou, taken to mean 'cow', the name perhaps meaning 'cow-place'.
Iter XII includes the data:
Nido (Neath)
Bomio m.p.xv
Iscae m.p.xxvii (Caerleon)
The total distance given for the journey from Neath to Caerleon is thus 6 Roman miles less than the true distance of 48 Roman miles. Now, if one travels 15 miles along that road from Neath one comes to the crossing of the Ewenny in the Bridgend area, a likely enough location for a Roman fort or settlement, and if one travels back along that road for 27 miles from Caerleon one comes to Cowbridge, where there is known to have been a Roman settlement. And the distance from Bridgend to Cowbridge is 6 Roman miles. It thus seems quite clear that there is simply a line missing from Iter XII. If we assume that the place at the crossing of the Ewenny was called Aventio (the then name of the river Ewenny), then Iter XII of the AI should read:
Nido
Aventio m.p.xv
Bomio m.p.vi
Iscae m.p.xxvii
The name Bomio, assuming the form is correct, includes the old-style topographical element bom, meaning 'high hill'. Now there is at Cowbridge a peninsula of raised ground bounded by the river Thaw on its N, W and S sides, the peninsula rising to a height of about 30 metres above the river. But immediately east of the peninsula the land rises to a height of some 120 metres above the river, so it is not clear to which hill the name Bomio refers, the peninsula itself or the high hill to the east of it. It is most likely, however, that Bomio was originally the name of one of the two hill-forts in the vicinity, most probably Caer Dynnaf on the other side of the river Thaw, and the name was simply transferred by the Romans to a new fort and/or settlement which they built at Cowbridge itself.
BORCOVICIO see VELURCION
